Fashion is Taking Shape: Understanding Clothing Preference Based on Body Shape From Online Sources

TL;DR

This paper introduces a dataset and a multi-photo method to analyze the correlation between body shape and clothing preferences, improving prediction accuracy over previous single-view or manual methods.

Contribution

It presents a new dataset and a multi-photo approach for estimating body shape and modeling clothing preferences, advancing large-scale understanding of fashion choices based on body shape.

Findings

Clothing categories are correlated with body shapes in real-world data.

Multi-photo shape estimation improves prediction of clothing preferences.

The proposed method outperforms single-view and manual annotation models.

Abstract

To study the correlation between clothing garments and body shape, we collected a new dataset (Fashion Takes Shape), which includes images of users with clothing category annotations. We employ our multi-photo approach to estimate body shapes of each user and build a conditional model of clothing categories given body-shape. We demonstrate that in real-world data, clothing categories and body-shapes are correlated and show that our multi-photo approach leads to a better predictive model for clothing categories compared to models based on single-view shape estimates or manually annotated body types. We see our method as the first step towards the large-scale understanding of clothing preferences from body shape.